Embed Size (px)
Citation preview
Visual Mapper with Speech RecognitionThe visualization and traversing of semantic knowledge graphs supported by speech recognition is a very innovative solution approach for working with knowledge assets. It can significantly improve e-learning systems. The PoolParty application "Visual Mapper" demonstrates how learning of the future can look like. The speech-recognition enhanced Visual Mapper is a JavaScript library o�ering an easy to configure graph displaying data relations in a user-friendly way. Combined with Google Chrome's Web Speech API, it allows you to browse through semantic data by voice.
Use Case
The challenge
Finding a graph type capable of interactively browsing through a SKOS thesaurus was a very unsatisfying task, the graphs were either confusing and really complicated to use or not able to display the complex data relations in a reasonable way. The graph closest to our needs was the Relation Browser created by Moritz Stefaner. As it is Flash-based, lacks configuration, is di�icult to style and requires all data as an XML-file instead of loading required data on the fly, this isn t a future-proof solution. A more flexible and configurable technology solution was in demand in order to build a foundation for innovative end-user applications.
The solution
The solution is a circle based graph type, where all the related concepts of the currently displayed URIs are grouped by their relation types. Each of these groups are ordered by the complexity of their concepts. The relation types of each of the concepts and
their percentage distribution are also visible without any additional operation required by the user. Every click on a concept loads its data on the fly from a specified data source and displays and replac-es the current chart data with the returned JSON data.
The results
The Visual Mapper is not only capable of displaying SKOS thesauri but any form relationship data, as long as every data point has its unique identifier. The Visual Mapper o�ers listeners for most of the actions happening inside the graph, a set of methods to control the graph through your own JavaScript ode and a wide variety of config-uration parameters to make the graph work and look like you want it to. The integration of speech recognition provided by Google Chrome into the Visual Mapper is currently still a at the start, but the faster and more powerful the Web Speech API gets, the more possibilities we have to make it interactively browse through datasets.
Technical architecture: Visual Mapper with Speech Recognition
Phone (EU): +43 1 4021235 Phone (USA): +1 (206) 683-8857Mail: [email protected]: +43 1 4021235 22Neubaugasse 1, 1070 Vienna, Austriawww.poolparty.biz
PoolParty is a semantic technology platform developed, owned and licensed by the Semantic Web Company. The company is also involved in international R&D projects, which have a continuous impact on the product development. The EU-based company has been a pionieer in the Semantic Web for over a decade.
Project insights How to develop semantic intelligence
STEP 1: DATA RESOURCES
Use an existing dataset (e.g. DBpedia) or create your own project using PoolParty Thesaurus Server.
STEP 2: PHP DATA FETCHING
O�er an endpoint providing the Visual Mapper with all the data it needs in the correct JSON format. Additional proper-ties are allowed and make it possible for you to add any additional behavior to the Visual Mapper using one of the provided JavaScript listeners.
STEP 3: VISUAL MAPPER
Create an HTML website for your project presentation, include all the required libraries and configure a new Visual Mapper instance for your needs.
REACH OUT TO THE DEVELOPER
Do you want to know more? Contact us!
MARTIN SCHAUER
DeveloperMail: [email protected]
TAKE A LOOK
http://visual.poolparty.biz
